### TABLE I PARAMETERS FOR THE POLICY GRADIENT ALGORITHM

2004

Cited by 5

### TABLE III SPECIALIZED LIKELIHOOD RATIO POLICY GRADIENT ESTIMATOR G(PO)MDP /POLICY GRADIENT WITH AN OPTIMAL BASELINE.

2006

Cited by 9

### TABLE III SPECIALIZED LIKELIHOOD RATIO POLICY GRADIENT ESTIMATOR G(PO)MDP /POLICY GRADIENT WITH AN OPTIMAL BASELINE.

### Table 1: Summary of the Bayesian policy gradient Models 1 and 2.

2007

"... In PAGE 4: ... 12, resulting in two distinct Bayesian models. Table1 summarizes the two models we use in this work. Our choice of Fisher-type kernels was motivated by the notion that a good representation should depend on the data generating process (see [13, 14] for a thorough discussion).... In PAGE 4: ... Our particular choices of linear and quadratic Fisher kernels were guided by the requirement that the posterior moments of the gradient be analytically tractable. In Table1 we made use of the following de nitions: F M = (f( 1; ); : : : ; f( M; )) N(0; KM), Y M = (y( 1); : : : ; y( M)) N(0; KM + 2I), UM = u( 1) ; u( 2) ; : : : ; u( M) , ZM = R r Pr( ; )kM( ) gt;d , and Z0 = RR k( ; 0)r Pr( ; )r Pr( 0; ) gt;d d 0. Finally, n is the number of policy parameters, and G = E u( )u( ) gt; is the Fisher information matrix.... ..."

Cited by 1

### Table 1: Summary of the Bayesian policy gradient Models 1 and 2.

2007

"... In PAGE 4: ... 12, resulting in two distinct Bayesian models. Table1 summarizes the two models we use in this work. Our choice of Fisher-type kernels was motivated by the notion that a good representation should depend on the data generating process (see [13, 14] for a thorough discussion).... In PAGE 4: ... Our particular choices of linear and quadratic Fisher kernels were guided by the requirement that the posterior moments of the gradient be analytically tractable. In Table1 we made use of the following de nitions: F M = (f( 1; ); : : : ; f( M; )) N(0; KM), Y M = (y( 1); : : : ; y( M)) N(0; KM + 2I), UM = u( 1) ; u( 2) ; : : : ; u( M) , ZM = R r Pr( ; )kM( ) gt;d , and Z0 = RR k( ; 0)r Pr( ; )r Pr( 0; ) gt;d d 0. Finally, n is the number of policy parameters, and G = E u( )u( ) gt; is the Fisher information matrix.... ..."

Cited by 1

### Table 2: Parameters for the policy gradient algorithm in the ball acquisition learning task

2007

Cited by 5

### Table 2: Parameters for the policy gradient algorithm in the ball acquisition learning task

### Table 3. Resistance to Cryptanalysis for Networks

1996

"... In PAGE 28: ... 8673. Summary of Results In Table3 , for SPNs of 8 rounds, we have summarized lower bounds on the values of 7868... In PAGE 29: ... Results are presented for networks using permutations from the set 5 and for networks using a linear transformation of the form of equation (23). Note that the analysis of Table3 is equally applicable to the decryption as well as the encryption network. (This is important since the decryption network may also be attacked using either cryptanalysis method.... ..."

Cited by 26

### Table 3. Resistance to Cryptanalysis for Networks

1994

"... In PAGE 28: ... a36a38a37 . Summary of Results In Table3 , for SPNs of 8 rounds, we have summarized lower bounds on the values of a108 a40a39... In PAGE 29: ... Results are presented for networks using permutations from the set a115 and for networks using a linear transformation of the form of equation (23). Note that the analysis of Table3 is equally applicable to the decryption as well as the encryption network. (This is important since the decryption network may also be attacked using either cryptanalysis method.... ..."

### Table 1. Summary of the cryptanalysis of DES.

Cited by 1